Model: SoLU Model: 10 Layers, 5120 Neurons per Layer

Dataset: 80% C4 (Web Text) and 20% Python Code

Neuron 1760 in Layer 1

Load this data into an Interactive Neuroscope

See Documentation here

Transformer Lens Loading: HookedTransformer.from_pretrained('solu-10l')

Text #0

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.6527. Min Act: -0.0003

Data Index: 287513 (C4 (Web Text))

Max Activating Token Index: 91

Text #1

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.5733. Min Act: -0.0003

Data Index: 38230 (C4 (Web Text))

Max Activating Token Index: 298

Text #2

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.5461. Min Act: -0.0003

Data Index: 807402 (C4 (Web Text))

Max Activating Token Index: 283

Text #3

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.5075. Min Act: -0.0003

Data Index: 661466 (C4 (Web Text))

Max Activating Token Index: 823

Text #4

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.5242. Min Act: -0.0003

Data Index: 529139 (C4 (Web Text))

Max Activating Token Index: 70

Text #5

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.5375. Min Act: -0.0003

Data Index: 293137 (C4 (Web Text))

Max Activating Token Index: 214

Text #6

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.5001. Min Act: -0.0003

Data Index: 781956 (C4 (Web Text))

Max Activating Token Index: 107

Text #7

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4625. Min Act: -0.0003

Data Index: 265755 (C4 (Web Text))

Max Activating Token Index: 379

Text #8

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4919. Min Act: -0.0003

Data Index: 1102271 (C4 (Web Text))

Max Activating Token Index: 844

Text #9

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4773. Min Act: -0.0003

Data Index: 1063799 (C4 (Web Text))

Max Activating Token Index: 38

Text #10

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4665. Min Act: -0.0002

Data Index: 1561669 (Python Code)

Max Activating Token Index: 837

Text #11

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4637. Min Act: -0.0002

Data Index: 1551952 (Python Code)

Max Activating Token Index: 650

Text #12

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4378. Min Act: -0.0003

Data Index: 1383669 (Python Code)

Max Activating Token Index: 504

Text #13

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4260. Min Act: -0.0003

Data Index: 529760 (C4 (Web Text))

Max Activating Token Index: 507

Text #14

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4355. Min Act: -0.0003

Data Index: 367544 (C4 (Web Text))

Max Activating Token Index: 53

Text #15

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4298. Min Act: -0.0002

Data Index: 1435528 (Python Code)

Max Activating Token Index: 713

Text #16

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4557. Min Act: -0.0003

Data Index: 55902 (C4 (Web Text))

Max Activating Token Index: 58

Text #17

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4274. Min Act: -0.0003

Data Index: 1304001 (C4 (Web Text))

Max Activating Token Index: 780

Text #18

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4385. Min Act: -0.0003

Data Index: 932977 (C4 (Web Text))

Max Activating Token Index: 395

Text #19

Max Range: 0.6527. Min Range: -0.6527

Max Act: 0.4006. Min Act: -0.0002

Data Index: 1549952 (Python Code)

Max Activating Token Index: 697